| With the rapid development of computer technology,industrial Internet realizes man-machine connection with the interconnection of intelligent machines.The emergence of industrial Internet makes the variety of data that need to be processed in the network,but also brings the exponential growth of the amount of data.In the industrial Internet,cloud computing,bandwidth,memory and other resources are limited,so a task scheduling strategy is urgently needed to solve the problem of profit optimization and load balancing.Firstly,the tasks are clustered and classified to avoid local optimization.Then the task scheduling problem is transformed into a bipartite graph matching problem,and the profit matrix of task scheduling is constructed.Finally,based on the matrix,the optimal bipartite graph matching is carried out to maximize the profit of task scheduling and load balancing,providing efficient and stable computing and resource management and control capabilities for industrial Internet.In view of the weak computing power of edge devices,which is not suitable for large-scale applications,a computing offload strategy is adopted to solve the energy consumption optimization problem under the delay constraint of edge devices.Firstly,the AOE network is used to model the dependence relationship between multi tasks.Then,the relaxation variables of AOE network are used to realize the delay constraint of unloading.Finally,with the energy consumption of edge devices as the optimization objective,a computing unloading strategy based on the minimum energy consumption of delay constraint is proposed to provide the edge end with the ability to reduce energy consumption and delay in industrial Internet.The simulation results show that the former algorithm has some advantages over the reference algorithm in terms of delay,reliability,profit and load balancing.The latter has some advantages in delay and energy consumption compared with local processing and full offload. |
PDF Full Text Request